5332
BIOCHEMISTRY
WONG, WINER, AND FREY
Stearn, A. E. (1949) Adc. Enzymol. Relat. Areas Mol. Biol. 9, 25-74. Tsong, T . Y., Baldwin, R. L., & Elson, E. L. (1972) Proc. Natl. Acad. Sci. U.S.A. 69, 1809-1812. Van Duuren, B. (1963) Chem. Rev. 63, 325-354. von Hippel, P. H . (1975) Protein-Ligand Interact., Proc. Symp., 1974, 452-47 1. von Hippel, P. H., & Schleich, T. (1969) in Structure and Stability of Biological Macromolecules (Timasheff, S. N., & Fasman, G. D., Eds.) pp 417-574, Marcel Dekker, New York. NY.
von Hippel, P. H., & Hamabata, A. (1973) J . Mechanochem. Cell Motil. 2, 127-138. von Hippel, P. H., Peticolas, V., Schack, L., & Karlson, L. (1973) Biochemistry 12, 1256-1264. Warren, J. C., & Cheatum, S. G. (1966) Biochemistry 5 , 1702-1 707. Weber, G., & Bablouzian, B. (1966) J . Biol. Chem. 241, 2558-2561. Wilkinson, G. N. (1961) Biochem. J . 80, 324-332. Zipp, A., & Kauzmann, W. (1973) Biochemistry 12, 421 7-4228.
p-( Bromoacetamido)phenyl Uridyl Pyrophosphate: An Active-Site-Directed Irreversible Inhibitor for Uridine Diphosphate Galactose 4-Epimeraset Yun-Hua Huang Wong, Fredric B. Winer, and Perry A. Frey*
ABSTRACT:
The synthesis of p-(bromoacetamido)phenyl uridyl pyrophosphate (BUP) is described. This compound is an active-site-directed irreversible inhibitor of Escherichia coli UDP-galactose 4-epimerase. The inactivation follows pseudofirst-order kinetics at pH 8.5 in nonnucleophilic buffers, and a saturation effect is seen in the pseudo-first-order rate constant as the concentration of BUP is increased. The half-saturation parameter for BUP in the inactivation is 0.21 f 0.02 mM,
which compares favorably with the inhibition constant of 0.3 f 0.05 mM for BUP acting as a competitive reversible inhibitor of the enzyme. The inactivation rate is slow, however, with a minimum half-time of 12 h at pH 8.5 and 27 OC. Both specific alkylation and nonspecific alkylation by BUP occur, but nonspecific alkylation is faster than the inactivation and the rate of inactivation correlates well with the rate of covalent incorporation of one molecule of [ 14C]BUP at the active site.
x e mechanism of the interconversionof UDP-galactose and UDP-glucose catalyzed by UDP-galactose 4-epimerase is known to involve the reversible formation of an epimeraseDPNH-UDP-4-ketohexose intermediate complex, eq 1 and 2 E.DPN+ + UDP-Gal + E.DPN+.UDP-Gal+ EeDPNH-UDP-4-ketohexose (1)
Scheme I
E-DPNH-UDP-4-ketohexose + E.DPN+.UDP-Glc E.DPN+ UDP-Glc (2) (Nelsestuen & Kirkwood, 1971; Maitra & Ankel, 1971; Wee & Frey, 1973; Adair et al., 1973). The interconversions of this central complex with the epimerase.DPN+.UDP-hexose complexes requires general acid-base catalysis, Le., general base catalyzed removal of the proton from the glycosyl C-4 hydroxyl groups of UDP-hexose substrates concomitant with their conversion to the ketonic intermediate and general acid catalyzed protonation of the ketonic oxygen in concert with its reduction to the galactosyl or glucosyl groups. Since the reduction process is the microscopic reverse of the oxidation process, or nearly the microscopic reverse, the general acid and general base functions can be expected to be performed by a single functional group which acts as a general base in the oxidation of glycosyl groups and as a general acid in the reduction of the 4-ketohexopyranosylgroup (Scheme I). While the identity of this group is not known, it is almost certainly present, because in the absence of such a group with a pK, in the near-physiological range the reversible redox process would involve the compulsory formation of high-energy inter-
+
+ From the Department of Chemistry, The Ohio State university, Columbus, Ohio 43210. Receiued March 6, 1979. Supported by Grant No. 13502 from the National Institute of Arthritis, Metabolism, and Digestive Diseases.
0006-2960/79/0418-5332$01 .OO/O
mediates, either the protonated 4-ketohexopyranosyl group or the alkoxide ion. A similar situation exists for all such reactions involving oxidation of an alcohol to a ketone, and general bases corresponding to that in Scheme I are known to be present in the active sites of pyridine nucleotide dependent dehydrogenases (Holbrook et al., 1975; Banaszak & Bradshaw, 1975). In these cases a histidyl residue appears to be appropriately situated to function as the prototropic catalyst. In order to identify this functional group, we have undertaken to synthesize an active-site-directed irreversible inhibitor which may alkylate the functional group when it is in its unprotonated, nucleophilic form. In the design of the inhibitor we have taken into account the binding properties of the active site of this enzyme as well as the degradation of the alkylated enzyme to an identifiable product. These considerations led us to synthesize p-(bromoacetamido)phenyl uridyl pyrophosphate (BUP),' 1.
' Abbreviations used: DPN, diphosphopyridine nucleotide; TPN, triphosphopyridine nucleotide; UMP, uridine 5'-monophosphate; Ans, 8anilino-1-naphthalenesulfonate; NUP, p-nitrophenyl uridyl pyrophosphate; AUP, p-aminophenyl uridyl pyrophosphate; BUP, p-(bromoacetamido)phenyl uridyl pyrophosphate; Bicine, N,N-bis(2-hydroxyethy1)glycine. 0 1979 American Chemical Society
P - ( B R O M O A C E T A M I D O ) P H E N Y LU R I D Y L P Y R O P H O S P H A T E 0 HN
I
4
I
OH OH
1
This compound contains the uridylpyrophosphoryl group, which is known to be the major contributor to the binding free energy for both substrates and the intermediate (Kang et al., 1975; Wong & Frey, 1977, 1978), so it should have good binding affinity for the active site. The p-(bromoacetamido)phenyl group is bonded to the nucleotide in place of the glycosyl group of the substrates. This substitution should not greatly affect the binding specificity or affinity because glycosyl groups of substrates, inhibitors, and intermediates contribute little to the binding free energy of uridine nucleotides at the active site (Wong & Frey, 1977). Thep-(bromoacetamido)phenyl group is a reasonably effective alkylating group introduced by Wilchek & Anfinsen (1969) in their chemical mapping studies of the active site of staphylococcal nuclease. It is a convenient group for such work because upon complete hydrolysis of an alkylated protein the alkylated a-amino acid should be a carboxymethyl derivative, which should be readily identified. In this paper we report on the synthesis of BUP, on the inactivation of UDP-galactose 4-epimerase by this compound, and on the stoichiometry of [I4C]BUPincorporation. BUP is here shown to be an active-site-directed alkylating agent for this enzyme, and in an accompanying article it is shown to alkylate the adenine ring of the DPN molecule that is present as a tightly bound coenzyme at the active site of this enzyme (Wong & Frey, 1979). Experimental Procedures Materials. UDP-galactose 4-epimerase was purified from a regulatory mutant of Escherichia coli K12 (ATCC 27797) by a slight modification of the procedure of Wilson & Hogness (1964). Galactose- 1-phosphate uridylyltransferase was purified from the same organism by Dr. Sue-Lein Lee Yang in this laboratory and kindly donated for this study. UDP-glucose dehydrogenase, phosphoglucomutase, glucose-6-phosphate dehydrogenase, nucleotide pyrophosphatase, and alkaline phosphatase were obtained from Sigma Chemical Co. UMP, TPN', glucose 1,6-diphosphate,glucose 1-phosphate, galactose 1-phosphate, UDP-glucose, UDP-galactose, and DPN+ were also obtained from Sigma. Other chemicals were obtained commercially and used without further purification except for Ans and pyridine. Ans was purchased from Sigma and twice recrystallized from water. Pyridine was redistilled and stored over KOH pellets. Assays. The concentration of UDP-galactose 4-epimerase used in the binding experiments was measured spectrophotometrically at 345 and 280 nm. The best measure of the content of active enzyme in any preparation of this enzyme has been found to be the content of DPN+ that is reducible by glucose in the presence of U M P (Kang et al., 1975). The increase in A345due to DPNH formation in each enzyme preparation used in this work upon adding glucose and UMP was used to calculate the active enzyme concentration, assuming an extinction coefficient of 6.2 X lo3. This was then correlated with the A280 of the enzyme and with the color development in the Lowry assay for protein. The concentration of inactivated
V O L . 18, NO. 24, 1979
5333
enzyme was measured by the Lowry method (Lowry et al., 1951). The assay method for UDP-galactose 4-epimerase was that of Wilson & Hogness (1964) except in the experiments to evaluate BUP as a competitive reversible inhibitor. A different assay method was used for this purpose. The amount of UDP-glucose produced from UDP-galactose in a timed incubation was measured as TPNH after adding galactose-lphosphate uridylyltransferase, galactose 1-phosphate, phosphoglucomutase, glucose-6-phosphate dehydrogenase, and TPN'. The reaction mixture contained initially, in 0.05 mL, 12.5 nmol of UDP-galactose, 0.125 pmol of potassium Bicinate buffer at pH 8.5, BUP at various concentrations, and less than 0.2 unit of enzyme. After 5 min at 27 "C the reaction was terminated by heating at 100 "C for 3 min. After cooling, we mixed the solution with 0.5 mL of another solution which contained 1 pmol of galactose 1-phosphate, 16 pmol of cysteine, 1.3 pmol of TPN', 4 nmol of glucose 1,6-diphosphate, 5 pmol of MgCI2, 5 pmol of NaF, 114 pmol of sodium Bicinate at pH 8.5,0.2 unit of glucose-6-phosphate dehydrogenase, 0.4 unit of phosphoglucomutase, and less than 0.01 unit of galactose- 1-phosphate uridylyltransferase (Wong & Frey, 1974). After 2 h at 27 "C the A3@was measured vs. a control solution to which no UDP-galactose 4-epimerase had been added. The unit of activity was the standard unit defined by Wilson & Hogness (1 964). Radioactivity was measured by liquid scintillation counting with a Packard Tri-Carb Model 3310 spectrometer. The scintillation medium contained 7 g of 2,5-diphenyloxazole,0.3 g of p-bis[2-(5-phenyloxazolyl)]benzene,and 100 g of naphthalene per L of 1,4-dioxane. A 1 .O-mL aqueous sample was mixed with 15 mL of this scintillator. Synthesis of BUP. Disodium p-nitrophenyl phosphate (4 mmol) was dissolved in 5 mL of H 2 0 and converted to the pyridinium salt by passage through a 2.5 X 26 cm column of Dowex 5OW-X12 cation-exchange resin in the pyridinium form. The salt was reduced to dryness by rotary evaporation in vacuo (bath temperature 30 "C). The residue was again evaporated several times after addition of 10-mL portions of anhydrous pyridine to remove the last traces of water. The residue was dissolved with 4 mmol of trioctylamine and 20 mL of anhydrous pyridine, evaporated to dryness, and dissolved in 20 mL of anhydrous pyridine containing 1.32 mmol of UMP-morpholidate. After evaporation 3 times with additions of anhydrous pyridine, the residual oil was dissolved in 5 mL of dry pyridine and placed at 45 "C for 2 days. The solvent was removed by rotary evaporation, and the residue was stirred with a small amount of water containing 600 mg of lithium acetate. This mixture was extracted several times with diethyl ether, and the aqueous layer was adjusted with HCl to pH 3.5. This solution was applied to a 4 X 40 cm column of DEAEcellulose in the C1- form which had been equilibrated with 1 mM HC1. The column was eluted with a linear gradient consisting of 1 L of 0.04 M LiCl in 1 mM HC1 and 1 L of 1.5 M LiCl in 1 mM HCl at a flow rate of 1 mL/min. Fractions were collected and monitored spectrophotometrically at 260 nm. Two minor bands of A260were eluted ahead of the major band. The first was pyridine and the second, eluted at the midpoint of the gradient, contained p-nitrophenyl phosphate and UMP. The major band eluted in the last 25% of the gradient was NUP. These fractions were pooled, adjusted to pH 5 with LiOH, and evaporated to dryness. The residue was dissolved in a minimal volume of cold methanol and precipitated by addition of 10 volumes of diethyl ether. The precipitate was collected by centrifugation, and the dis-
5334
BIOCH EM ISTRY
solution with methanol and precipitation with ether were repeated until the supernatant fluid was free of Cl-. Li2NUP was obtained in a yield of 507-598 mg (7344%). The uracil/phosphate/p-nitrophenolratio was found to be 1.0:2.1:1 .O upon degradation with nucleotide pyrophosphatase and alkaline phosphatase at pH 8, followed by spectrophotometric analysis at 260 nm for uracil and at 400 nm for p-nitrophenol and colorimetric analysis for phosphate. Pure Li2NUP exhibited an extinction coefficient at pH 7.5 of 11 X lo3 at its absorption maximum of 268 nm. N U P was hydrogenated to AUP in 1.5 h at 32 psi and room temperature in a Parr hydrogenation apparatus. The hydrogenation mixture consisted of 100 mg of NUP dissolved in 100 mL of 50% aqueous methanol (v/v) and mixed with 32 mg of 10% Pd on charcoal catalyst. After filtration of the catalyst, the solution was evaporated to dryness. The residue was dissolved in a minimal volume of methanol and precipitated with ether. The precipitate was dissolved in a minimum of methanol and again precipitated with ether. AUP obtained in over 90% yield was homogeneous upon paper chromatography and exhibited an extinction coefficient of 8.3 X lo3 at pH 7.5 at its absorption maximum of 262 nm. N-Hydroxysuccinimide bromoacetate and N-hydroxysuccinimide [ 2-14C]bromoacetate, prepared according to Santi & Cunnion (1974), were used to bromoacetylate AUP to BUP. AUP (1 50 mg, 0.3 mmol) was dissolved in 1.5 mL of water and combined with 4.2 mL of redistilled dioxane. This was combined with 177 mg (0.75 mmol) of N-hydroxysuccinimide bromoacetate, and the solution was gently stirred for 3-4 h a t room temperature. One volume of water was added, and the resulting solution was extracted several times with equal volumes of ether. The aqueous layer was adjusted with HC1 to pH 3.5 and applied to a 1.5 X 42 cm column of DEAESephadex A-25 in the Br- form. The column was eluted with a linear gradient consisting of 250 mL of 0.05 M LiBr in 1 m M HC1 and 250 mL of 0.35 M LiBr in 1 mM HCl. Fractions of 4.0-mL volume were collected at 6-min intervals at 4 OC. Two very minor bands were eluted near the midpoint of the gradient, and BUP was eluted as the major band in the last 25% of the gradient. The fractions containing BUP were pooled, adjusted to pH 5 with LiOH, and evaporated to dryness. The residue was dissolved in methanol, and Li2BUPwas precipitated by addition of diethyl ether, as described for NUP. Analytically pure dilithium BUP was obtained in 50% yield. Anal. Calcd: C, 29.93; H, 3.54; N, 6.16; Br, 11.7. Found: C, 29.69; H, 3.64; N , 6.05; Br, 11-41, The compound exhibited an extinction coefficient of 1.6 X lo4 at pH 7.0 at its absorption maximum of 258 nm. [bromoa~eryl-2-*~C]BUP was prepared by an otherwise identical but scaled-down procedure. Sodium [2-I4C]bromoacetate was obtained from Amersham/Searle. The specific activity of the ['4C]BUP was 1 X lo6 cpm/pmol. Results and Discussion Synthesis of BUP. The synthetic procedure we have developed for preparing BUP begins with the reaction of p nitrophenyl phosphate with the trioctylammonium salt of UMP-morpholidate in pyridine to produce NUP. This reaction is carried out by the procedure introduced by Moffatt & Khorana (1958) which is in general use for the synthesis of nucleotide sugars (Moffatt, 1966). N U P is reduced to AUP by catalytic hydrogenation of the nitro group, and AUP is bromoacetylated to BUP by reaction with bromoacety1-Nhydroxysuccinimide. The latter two steps are adapted from the analogous reactions used by Wilchek & Anfinsen (1 969) to synthesize p-(bromoacetamido)phenyl nucleotide analogues
WONG, WINER, AND FREY
h(nrn)
Ultraviolet spectra of BUP and synthetic intermediates. The ultraviolet spectra of 0.055 mM BUP in 0.1 M KPi buffer at pH 7.0 and of 0.050 mM N U P and AUP in 0.1 M KP,buffer at pH 7.5 are shown. FIGURE 1:
of staphylococcal nuclease substrates. BUP is obtained analytically pure by the procedure described under Experimental Procedures. The ultraviolet spectra of NUP, AUP, and BUP are given in Figure 1 in support of the structure assigned to BUP. The structure of NUP is defined by the synthetic procedure. The only functional group that can be hydrogenated under the conditions used is the nitro group, since hydrogenation of uracil requires high pressure and a nickel catalyst, so the structure of AUP is unambiguous. The ultraviolet spectrum of N U P in Figure 1 is approximately the sum of the spectral components, whereas AUP is clearly hypochromic with an extinction coefficient at 260 nm that is smaller than that of UMP itself. It appears that the delocalization of the electron pair on the amino group enriches the electron density in the phenyl ring enough to promote a stacking interaction with the uracil ring. Upon bromoacetylation, hypochromicity is relieved in BUP, presumably because the electron pair on the amide nitrogen is now less delocalized into the phenyl ring because of its delocalization into the bromoacetyl group. This suggests that it is the anilino nitrogen that is bromoacetylated. Inactivation of UDP-Galactose 4-Epimerase by BUP. An active-site-directed irreversible inhibitor inactivates its target enzyme by forming a specific association complex at the active site and then chemically modifying one or more functional groups of the protein at this site, usually by forming a covalent bond to the group. The kinetics for such inactivation is known by the work of Fahrney & Gold (1963) on inactivators of a-chymotrypsin to be described by eq 3 and 4. In eq 4,kow
(4)
is the pseudo-first-order rate constant for inactivation of the enzyme E by a substantial excess of the inhibitor I. The saturation parameter K is ( k 2+ k , ) / k l when the appropriate approximation is the steady-state approximation. When k3